Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    code snippets

    np_max-act · gemini-2.0-flash

    The neuron fires on code identifiers—variable, type, and function names in source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	AL
    -0.06
     Linden
    -0.06
    енням
    -0.06
    ungkin
    -0.06
    _|
    -0.06
    icular
    -0.06
    ريب
    -0.06
    ratulations
    -0.06
     Digest
    -0.06
     Happ
    -0.06
    POSITIVE LOGITS
     ports
    0.06
    víc
    0.06
    diğini
    0.06
    "%(
    0.06
     uranus
    0.06
    apollo
    0.06
    .Local
    0.06
     gbc
    0.06
    -over
    0.06
    Loss
    0.06
    Act Density 0.030%

    No Known Activations